Blundell and Bond(1998) panel system GMM

[banya56]

Can we extend the allerano-bond model (1991) into Blundell and Bond(1998)'s system GMM ?

[randerson]

I wondered if anyone has worked on this. The comparable Stata extension is described at: http://www.stata.com/manuals14/xtxtdpdsys.pdf. It appears in Stata they simply expand the instrument list but the discussion in Baltagi (2013), sec 8.5, does not make this seem like the obvious way to proceed: "...the use of an extended system GMM estimator that uses lagged differences of y(i,t) as instruments for equations in levels, in addition to lagged levels of y(i,t) as instruments for equations in first differences." This also appears to be sec 11.4 in Wooldridge (2010) but he does not link the discussions of estimators to published literature. This seems straightforward, perhaps an extension on UG-423?

The TextbookExample bpanel08-8-1.rpf shows the various estimators on page 177 (Table 1) of Baltagi, 5th ed, but it omits the "System GMM" Blundell-Bond estimator (the example is written for the 4th ed, but the context is clear). Perhaps it is easy to add that estimator to the end of that example for us not completely familiar with alternative panel methods...

Thanks

Richard

[BinhPham]

I guess System GMM is actually in the wish-list for next release of RATS . We all need it for micro econometrics. Please add it asap!

[TomDoan]

Note that "system GMM" in general is done fairly easily with NLSYSTEM. That does GMM as described in Hansen with the same instruments for each condition (Kroneker product of conditions x instruments). Blundell/Arellano/Bond estimators are much more complicated because there are different instruments for each equation (which isn't that big a deal---NLSYSTEM has a mask option to handle that), and also different instruments for each data point.

[shimarats]

hi dear tom

i foundi several code from gmm panel but i want Blundell & Bond (1998) panel system code from you . please give me this type of panel gmm in rats .thanksssssssssssssssssssss

[shengyang]

Hi Tom,

Would you give an example on how to set up separate sets of instrumental variables for simultaneous equations. Thank you so much for your help!

[shimarats]

HI DR TOM
ifollow your panel gmm code that i put bellow:

open data cornwell&rupert.xls

calendar(panelobs=7,a) 1976

data(format=xls,org=columns) 1//1976:01 595//1982:01 exper wks occ ind $

south smsa ms fem union ed blk lwage

*

set expersq = exper^2

*

set dwks = wks-wks{1}

set dlwage = lwage-lwage{1}

set dunion = union-union{1}

set docc = occ-occ{1}

set dexper = exper-exper{1}

*

* IV on first differences

*

@ablags wks abwks

instruments lwage{0 1} union{0 1} occ{0 1} exper{0 1} abwks

linreg(inst) dwks

# dlwage dunion docc dexper dwks{1}

iam using rats 8.0 in this analysis but i got different error.special in @ablags wks abwks . give me this error "# CP18. ABLAGS is not the Name of a PROCEDURE. (Did you forget to SOURCE?)
>>>>@ablags <<<<
please help me.ia waiting for your reply me.


thanksssssssssssssssssssssssssssss

[TomDoan]

HI DR TOM

iam using rats 8.0 in this analysis but i got different error.special in @ablags wks abwks . give me this error "# CP18. ABLAGS is not the Name of a PROCEDURE. (Did you forget to SOURCE?)
>>>>@ablags <<<<
please help me.ia waiting for your reply me.


thanksssssssssssssssssssssssssssss

You must have your procedure directory pointing to a location that doesn't exist any longer. Change that on the directories tab of the Preferences.